Short-term changes in the numerical density of synapses in the intermediate and medial hyperstriatum ventrale following one-trial passive avoidance training in the chick.

Previous ultrastructural studies using stereological counting techniques, based on assumptions regarding shape, size, and orientation of synapses, have suggested synaptic remodeling occurred at least 24 hr after one-trial passive avoidance training in day-old chicks. The present study estimates the mean synaptic density (Nv syn) in a region of the chick forebrain known to be involved in memory formation, the intermediate and medial hyperstriatum ventrale (IMHV), 1 and 24 hr following one-trial passive avoidance training. A stereological technique, the "disector," that makes no assumptions about size, shape, and orientation of synapses was used in the synaptic analyses. The density of axospinous synapses increased by approximately 77% at 1 hr posttraining in the right IMHV of chicks (M-trained) that learned to avoid a bitter-tasting bead, compared to those (W-trained controls) that peck a water-coated bead. A measure of the postsynaptic density size, the mean projected height of synapses (H), was 57% smaller 1 hr posttraining in the right IMHV of M-trained chicks. These differences were not found at 24 hr posttraining. We suggest that structural modification of synapses may be a key part of the processes involved in short-term memory formation.